/[webpac]/trunk/all2xml.pl
This is repository of my old source code which isn't updated any more. Go to git.rot13.org for current projects!
ViewVC logotype

Annotation of /trunk/all2xml.pl

Parent Directory Parent Directory | Revision Log Revision Log

Revision 678 - (hide annotations)
Sun Feb 27 23:07:35 2005 UTC (19 years, 1 month ago) by dpavlin
File MIME type: text/plain
File size: 28947 byte(s) 
Experimental support for dBase .dbf files. Usege like this in all2xml.conf:

[hda]
       dbf_file=/data/drustvene/hda/ISO.DBF
       type=dbf
       dbf_codepage=cp852
       dbf_mapping=<<_END_OF_MAP_
ID_BROJ                001
ISBN_BROJ      010
SKUPINA1       200
SKUPINA2       205
SKUPINA4       210
SKUPINA5       215
SKUPINA6       225
SKUPINA7       300
ANOTACIJA      330
PREDMET1       610
PREDMET2       610
PREDMET3       510
UDK            675
REDALICA       700
SIGNATURA      990
_END_OF_MAP_

dbf type will use <isis> tag in import_xml and dbf_codepage will
override codepage specified in import_xml file.

Small code refactoring.
1 dpavlin 1 #!/usr/bin/perl -w
2    
3     use strict;
4 dpavlin 632 use Biblio::Isis;
5 dpavlin 1 use Getopt::Std;
6     use Data::Dumper;
7     use XML::Simple;
8 dpavlin 10 use Text::Iconv;
9 dpavlin 13 use Config::IniFiles;
10 dpavlin 40 use Encode;
11 dpavlin 177 #use GDBM_File;
12     use Fcntl; # for O_RDWR
13     use TDB_File;
14 dpavlin 678 use Carp;
15 dpavlin 1
16 dpavlin 10 $|=1;
17 dpavlin 9
18 dpavlin 13 my $config_file = $0;
19     $config_file =~ s/\.pl$/.conf/;
20 dpavlin 320 $config_file = $ARGV[0] if ($ARGV[0] && -f $ARGV[0]);
21 dpavlin 13 die "FATAL: can't find configuration file '$config_file'" if (! -e $config_file);
22    
23 dpavlin 10 my $config;
24    
25 dpavlin 58 #use index_DBI; # default DBI module for index
26 dpavlin 641 #use index_DBI_cache; # faster DBI module using memory cache
27 dpavlin 642 use index_DBI_filter; # filter support for indexes
28 dpavlin 50 my $index;
29 dpavlin 10
30 dpavlin 1 my %opts;
31    
32 dpavlin 7 # usage:
33     # -d directory name
34     # -m multiple directories
35     # -q quiet
36     # -s run swish
37 dpavlin 1
38 dpavlin 7 getopts('d:m:qs', \%opts);
39    
40 dpavlin 43 my $path; # this is name of database
41 dpavlin 1
42 dpavlin 57 Text::Iconv->raise_error(0); # Conversion errors don't raise exceptions
43 dpavlin 1
44 dpavlin 40 # this is encoding of all files on disk, including import_xml/*.xml file and
45     # filter/*.pm files! It will be used to store strings in perl internally!
46     my $codepage = 'ISO-8859-2';
47 dpavlin 1
48 dpavlin 40 my $utf2cp = Text::Iconv->new('UTF-8',$codepage);
49     # this function will convert data from XML files to local encoding
50     sub x {
51     return $utf2cp->convert($_[0]);
52     }
53 dpavlin 3
54 dpavlin 54 # decode isis/excel or other import codepage
55     my $import2cp;
56 dpavlin 10
57 dpavlin 40 # outgoing xml must be in UTF-8
58     my $cp2utf = Text::Iconv->new($codepage,'UTF-8');
59 dpavlin 29
60 dpavlin 54 # mapping between data type and tag which specify
61     # format in XML file
62     my %type2tag = (
63     'isis' => 'isis',
64 dpavlin 62 'excel' => 'column',
65     'marc' => 'marc',
66 dpavlin 678 'feed' => 'feed',
67     'dbf' => 'isis', # special case, re-use isis import_xml
68 dpavlin 54 );
69 dpavlin 3
70 dpavlin 170 my $cache; # for cacheing
71    
72 dpavlin 177 # lookup hash (tied to file)
73     my %lhash;
74     # this option will cache all lookup entries in memory.
75     # if you are tight on memory, turn this off
76     my $use_lhash_cache = 1;
77    
78 dpavlin 195 my $last_field_name; # cache to prevent repeated fields
79    
80 dpavlin 54 sub data2xml {
81    
82 dpavlin 10 use xmlify;
83    
84 dpavlin 54 my $type = shift @_;
85 dpavlin 3 my $row = shift @_;
86 dpavlin 13 my $add_xml = shift @_;
87 dpavlin 59 # needed to read values from configuration file
88     my $cfg = shift @_;
89     my $database = shift @_;
90 dpavlin 3
91     my $xml;
92    
93 dpavlin 10 use parse_format;
94 dpavlin 3
95 dpavlin 13 my $html = ""; # html formatted display output
96 dpavlin 10
97 dpavlin 13 my %field_usage; # counter for usage of each field
98    
99 dpavlin 32 # sort subrouting using order="" attribute
100     sub by_order {
101 dpavlin 98 my $va = $config->{indexer}->{$a}->{order} ||
102     $config->{indexer}->{$a};
103     my $vb = $config->{indexer}->{$b}->{order} ||
104     $config->{indexer}->{$b};
105 dpavlin 29
106 dpavlin 98 return $va <=> $vb;
107 dpavlin 32 }
108 dpavlin 3
109 dpavlin 170 my @sorted_tags;
110 dpavlin 180 if ($cache->{tags_by_order}) {
111     @sorted_tags = @{$cache->{tags_by_order}};
112 dpavlin 170 } else {
113     @sorted_tags = sort by_order keys %{$config->{indexer}};
114 dpavlin 180 $cache->{tags_by_order} = \@sorted_tags;
115 dpavlin 170 }
116 dpavlin 32
117 dpavlin 678 if (! @sorted_tags) {
118     print STDERR "WARNING: no tags for this type found in import_xml file!\n";
119     }
120    
121 dpavlin 177 # lookup key
122     my $lookup_key;
123    
124 dpavlin 178 # cache for field in pages
125 dpavlin 180 delete $cache->{display_data};
126     delete $cache->{swish_data};
127     delete $cache->{swish_exact_data};
128 dpavlin 182 delete $cache->{index_data};
129 dpavlin 188 delete $cache->{index_delimiter};
130 dpavlin 178 my @page_fields; # names of fields
131    
132    
133     # subs used to produce output
134    
135     sub get_field_name($$$) {
136     my ($config,$field,$field_usage) = @_;
137    
138     # find field name (signular, plural)
139     my $field_name = "";
140     if ($config->{indexer}->{$field}->{name_singular} && $field_usage == 1) {
141     $field_name = $config->{indexer}->{$field}->{name_singular};
142     } elsif ($config->{indexer}->{$field}->{name_plural}) {
143     $field_name = $config->{indexer}->{$field}->{name_plural};
144     } elsif ($config->{indexer}->{$field}->{name}) {
145     $field_name = $config->{indexer}->{$field}->{name};
146     } else {
147     print STDERR "WARNING: field '$field' doesn't have 'name' attribute!";
148     }
149 dpavlin 599
150 dpavlin 178 if ($field_name) {
151 dpavlin 599 $field_name = x($field_name);
152 dpavlin 195 if (! $last_field_name) {
153 dpavlin 599 $last_field_name = $field_name;
154 dpavlin 195 return $last_field_name;
155     } elsif ($field_name ne $last_field_name) {
156 dpavlin 599 $last_field_name = $field_name;
157 dpavlin 195 return $last_field_name;
158     }
159 dpavlin 178 }
160     }
161    
162 dpavlin 279
163     # init variables for different types
164 dpavlin 274 sub init_visible_type($) {
165     my $type = shift;
166 dpavlin 178
167 dpavlin 274 # swish, swish_exact, display, index, index_lookup
168     # swish and display defaults
169     my ($s,$se,$d,$i,$il) = (1,0,1,0,0);
170     if (lc($type) eq "display") {
171     $s = 0;
172     } elsif (lc($type) eq "swish") {
173     $d = 0;
174     } elsif (lc($type) eq "index") {
175     ($s,$se,$d,$i) = (0,1,0,1);
176     } elsif (lc($type) eq "swish_exact") {
177     ($s,$se,$d,$i) = (0,1,0,0);
178     } elsif (lc($type) =~ /^lookup/) {
179     ($s,$se,$d,$i,$il) = (0,1,0,0,1);
180 dpavlin 333 } elsif ($type) {
181     print STDERR "WARNING: unknown type: $type\n";
182 dpavlin 274 }
183     return ($s,$se,$d,$i,$il);
184     }
185    
186 dpavlin 279
187     # convert
188     #
189     # <tag>
190     # <delimiter>, </delimiter>
191     # <value>200a</value>
192     # </tag>
193     #
194     # to
195     #
196     # <tag delimiter=", ">200a</tag>
197     #
198     # but without loosing spaces in delimiter (becasue
199     # new XML::Simple strips spaces in attribute values
200     # as defined in XML specification)
201     #
202     sub unroll_x($) {
203     my $x = shift;
204    
205     if (defined $x->{value}) {
206     my ($v,$d) = ($x->{value}->{content}, $x->{delimiter}->{content});
207     delete $x->{value};
208     delete $x->{delimiter};
209     $x->{content} = $v;
210     $x->{delimiter} = $d;
211     }
212     return $x;
213     }
214    
215 dpavlin 178 # begin real work: go field by field
216 dpavlin 170 foreach my $field (@sorted_tags) {
217    
218 dpavlin 40 $field=x($field);
219 dpavlin 13 $field_usage{$field}++;
220    
221 dpavlin 10 my $swish_data = "";
222 dpavlin 163 my $swish_exact_data = "";
223 dpavlin 3 my $display_data = "";
224 dpavlin 182 my @index_data;
225 dpavlin 35 my $line_delimiter;
226 dpavlin 3
227 dpavlin 34 my ($swish,$display);
228    
229 dpavlin 54 my $tag = $type2tag{$type} || die "can't find which tag to use for type $type";
230 dpavlin 178
231     # is this field page-by-page?
232     my $iterate_by_page = $config->{indexer}->{$field}->{iterate_by_page};
233     push @page_fields,$field if ($iterate_by_page);
234     my %page_max = ();
235     # default line_delimiter if using
236     my $page_line_delimiter = $config->{indexer}->{$field}->{page_line_delimiter} || '<br/>';
237 dpavlin 188 $cache->{index_delimiter}->{$field} = $config->{indexer}->{$field}->{index_delimiter};
238 dpavlin 178
239 dpavlin 195 my $format_name = $config->{indexer}->{$field}->{format_name};
240     my $format_delimiter = $config->{indexer}->{$field}->{format_delimiter};
241     if ($format_name && $format_delimiter) {
242     $cache->{format}->{$field}->{format_name} = $format_name;
243     $cache->{format}->{$field}->{format_delimiter} = $format_delimiter;
244     }
245    
246 dpavlin 54 foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) {
247 dpavlin 3
248 dpavlin 279 $x = unroll_x($x);
249    
250 dpavlin 40 my $format = x($x->{content});
251     my $delimiter = x($x->{delimiter}) || ' ';
252 dpavlin 3
253 dpavlin 178 my $repeat_off = 0; # init repeatable offset
254 dpavlin 29
255 dpavlin 274 my ($s,$se,$d,$i,$il) = init_visible_type($x->{type});
256 dpavlin 29
257 dpavlin 35 # what will separate last line from this one?
258 dpavlin 182 if ($display_data && $x->{append}) {
259 dpavlin 290 $line_delimiter = $delimiter;
260 dpavlin 35 } elsif ($display_data) {
261 dpavlin 34 $line_delimiter = '<br/>';
262     }
263 dpavlin 29
264 dpavlin 34 # init vars so that we go into while...
265     ($swish,$display) = (1,1);
266 dpavlin 29
267 dpavlin 195 sub mkformat($$) {
268 dpavlin 153 my $x = shift || die "mkformat needs tag reference";
269     my $data = shift || return;
270 dpavlin 263 my $format_name = x($x->{format_name}) || return $data;
271 dpavlin 153 my $fmt = x($config->{format}->{$format_name}->{content}) || die "<format name=\"$format_name\"> is not defined!";
272     my $format_delimiter = x($x->{format_delimiter});
273     my @data;
274     if ($format_delimiter) {
275     @data = split(/$format_delimiter/,$data);
276     } else {
277     push @data,$data;
278     }
279    
280     if ($fmt) {
281     my $nr = scalar $fmt =~ s/%s/%s/g;
282     if (($#data+1) == $nr) {
283 dpavlin 263 return sprintf($fmt,@data);
284 dpavlin 153 } else {
285 dpavlin 207 #print STDERR "mkformat: [$data] can't be split on [$format_delimiter] to $nr fields!\n";
286 dpavlin 263 return $data;
287 dpavlin 153 }
288     } else {
289     print STDERR "usage of link '$format_name' without defined format (<link> tag)\n";
290     }
291     }
292    
293 dpavlin 90 # while because of repeatable fields
294     while ($swish || $display) {
295 dpavlin 178 my $page = $repeat_off;
296     $page_max{$field} = $page if ($iterate_by_page && $page > ($page_max{$field} || 0));
297 dpavlin 54 ($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp);
298 dpavlin 90 if ($repeat_off > 1000) {
299     print STDERR "loop (more than 1000 repeatable fields) deteced in $row, $format\n";
300     last;
301     }
302 dpavlin 177
303     # is this field is lookup?
304     if ($display && $x->{lookup}) {
305 dpavlin 178 my $null = "<!-- null -->";
306 dpavlin 177 if ($use_lhash_cache) {
307     if (!defined($cache->{lhash}->{$display})) {
308     my $new_display = $lhash{$display};
309 dpavlin 178 if (defined($new_display)) {
310 dpavlin 177 #print STDERR "lookup cache store '$display' = '$new_display'\n";
311     $display = $new_display;
312     $cache->{lhash}->{$display} = $new_display;
313     } else {
314 dpavlin 255 # print STDERR "WARNING: lookup for '$display' didn't find anything.\n";
315 dpavlin 177 $display = "";
316 dpavlin 178 $cache->{lhash}->{$display} = $null;
317 dpavlin 177 }
318     } else {
319     $display = $cache->{lhash}->{$display};
320     }
321     } else {
322 dpavlin 178 $display = $lhash{$display} || $null;
323 dpavlin 177 }
324     }
325    
326 dpavlin 29 # filter="name" ; filter this field through
327     # filter/[name].pm
328     my $filter = $x->{filter};
329 dpavlin 170 if ($filter && !$cache->{filter_loaded}->{$filter}) {
330 dpavlin 29 require "filter/".$filter.".pm";
331 dpavlin 170 $cache->{filter_loaded}->{$filter}++;
332 dpavlin 20 }
333 dpavlin 29 # type="swish" ; field for swish
334 dpavlin 163 if ($swish) {
335 dpavlin 259 my $tmp = $swish;
336 dpavlin 163 if ($filter && ($s || $se)) {
337 dpavlin 29 no strict 'refs';
338 dpavlin 259 $tmp = join(" ",&$filter($tmp)) if ($s || $se);
339 dpavlin 29 }
340 dpavlin 234
341 dpavlin 255 $swish_data .= $tmp if ($s && $tmp);
342 dpavlin 256 $swish_exact_data .= "xxbxx $tmp xxexx " if ($tmp && $tmp ne "" && $se);
343 dpavlin 29 }
344 dpavlin 17
345 dpavlin 29 # type="display" ; field for display
346     if ($d && $display) {
347 dpavlin 207 my $ldel = $delimiter;
348 dpavlin 35 if ($line_delimiter && $display_data) {
349 dpavlin 207 $ldel = $line_delimiter;
350 dpavlin 35 }
351 dpavlin 29 if ($filter) {
352     no strict 'refs';
353 dpavlin 207 my @arr;
354     foreach my $tmp (&$filter($display)) {
355     my $tmp2 = mkformat($x,$tmp);
356     push @arr,$tmp2 if ($tmp2);
357 dpavlin 138 }
358 dpavlin 207 $display_data .= $ldel if ($display_data && @arr);
359     $display_data .= join($delimiter,@arr);
360 dpavlin 29 } else {
361 dpavlin 207 $display_data .= $ldel if ($display_data);
362     my $tmp = mkformat($x,$display);
363 dpavlin 224 $display_data .= $tmp if ($tmp);
364 dpavlin 29 }
365 dpavlin 20 }
366 dpavlin 29
367     # type="index" ; insert into index
368 dpavlin 182 my $idisplay;
369 dpavlin 29 if ($i && $display) {
370 dpavlin 182 $idisplay = $display;
371 dpavlin 177 if ($filter) {
372     no strict 'refs';
373 dpavlin 182 $idisplay = &$filter($idisplay);
374 dpavlin 177 }
375 dpavlin 255 push @index_data, $idisplay if ($idisplay && !$iterate_by_page);
376 dpavlin 97 }
377    
378 dpavlin 177 # store fields in lookup
379     if ($il && $display) {
380     if (lc($x->{type}) eq "lookup_key") {
381     if ($lookup_key) {
382     print STDERR "WARNING: try to redefine lookup_key (keys shouldn't be repeatable fields!)";
383     } else {
384 dpavlin 233 if ($filter) {
385     no strict 'refs';
386     $lookup_key = &$filter($display);
387     } else {
388     $lookup_key = $display;
389     }
390 dpavlin 177 }
391     } elsif (lc($x->{type}) eq "lookup_val") {
392     if ($lookup_key) {
393 dpavlin 233 if ($filter) {
394     no strict 'refs';
395     $lhash{$lookup_key} = &$filter($display);
396     } else {
397     $lhash{$lookup_key} = $display;
398     }
399 dpavlin 177 } else {
400     print STDERR "WARNING: no lookup_key defined for '$display'?";
401     }
402 dpavlin 20 }
403 dpavlin 178
404 dpavlin 20 }
405 dpavlin 178
406     # store data for page-by-page repeatable fields
407     if ($iterate_by_page) {
408     sub iterate_fld($$$$$$) {
409     my ($cache,$what,$field,$page,$data,$append) = @_;
410     return if (!$data);
411 dpavlin 182
412     my $ldel = $page_line_delimiter;
413     $ldel = " " if ($append);
414     #print STDERR "line delimiter: ",Dumper($ldel) if ($ldel);
415 dpavlin 178 if (! $cache->{$what}->{$field}->[$page]) {
416     $cache->{$what}->{$field}->[$page] = $data;
417     } else {
418 dpavlin 182 $cache->{$what}->{$field}->[$page] .= $ldel.$data;
419 dpavlin 178 }
420     }
421    
422     if ($display_data) {
423     iterate_fld($cache,'display_data',$field,$page,$display_data,$x->{append});
424     }
425     $display_data = "";
426     if ($swish_data) {
427     iterate_fld($cache,'swish_data',$field,$page,$swish_data,$x->{append});
428     $swish_data = "";
429     }
430     if ($swish_exact_data) {
431     iterate_fld($cache,'swish_exact_data',$field,$page,$swish_exact_data,$x->{append});
432     $swish_exact_data = "";
433     }
434 dpavlin 182
435     if ($idisplay) {
436     my $ldel=$page_line_delimiter;
437     my @index_data;
438     if ($cache->{index_data}->{$field}->[$page]) {
439    
440     @index_data = @{$cache->{index_data}->{$field}->[$page]};
441     }
442     if ($x->{append}) {
443 dpavlin 195 if (@index_data) {
444     $index_data[$#index_data] .= $idisplay;
445     } else {
446     push @index_data, $idisplay;
447     }
448 dpavlin 182 } else {
449     push @index_data, $idisplay;
450     }
451     $idisplay = "";
452     @{$cache->{index_data}->{$field}->[$page]} = @index_data;
453     }
454 dpavlin 178 }
455 dpavlin 17 }
456 dpavlin 177
457 dpavlin 182 if (! $iterate_by_page) {
458 dpavlin 188 my $idel = $x->{index_delimiter};
459 dpavlin 182 # fill data in index
460 dpavlin 188 foreach my $tmp (@index_data) {
461     my $i = $d = $tmp;
462     if ($idel && $tmp =~ m/$idel/) {
463     ($i,$d) = split(/$idel/,$tmp);
464     }
465     $index->insert($field, $i, $d, $path);
466 dpavlin 182 }
467     @index_data = ();
468 dpavlin 177 }
469 dpavlin 3 }
470 dpavlin 9
471 dpavlin 59 # now try to parse variables from configuration file
472     foreach my $x (@{$config->{indexer}->{$field}->{'config'}}) {
473 dpavlin 13
474 dpavlin 279 $x = unroll_x($x);
475    
476 dpavlin 62 my $delimiter = x($x->{delimiter}) || ' ';
477 dpavlin 59 my $val = $cfg->val($database, x($x->{content}));
478    
479 dpavlin 274 # FIXME index_lookup is not supported!
480     my ($s,$se,$d,$i,$il) = init_visible_type($x->{type});
481 dpavlin 59
482     if ($val) {
483 dpavlin 62 $display_data .= $delimiter.$val if ($d);
484 dpavlin 274 $swish_data .= " ".$val if ($s);
485 dpavlin 59 $index->insert($field, $val, $path) if ($i);
486     }
487    
488 dpavlin 178 if ($iterate_by_page) {
489     # FIXME data from config tag will appear just
490     # on first page!!!
491     my $page = 0;
492     if ($display_data) {
493     $cache->{display_data}->{$field}->[$page] = $display_data;
494     $display_data = "";
495     }
496     if ($swish_data) {
497     $cache->{swish_data}->{$field}->[$page] = $swish_data;
498     $swish_data = "";
499     }
500     if ($swish_exact_data) {
501     $cache->{swish_exact_data}->{$field}->[$page] = $swish_exact_data;
502     $swish_exact_data = "";
503     }
504     }
505 dpavlin 59 }
506    
507 dpavlin 178 # save data page-by-page
508     foreach my $field (@page_fields) {
509     my $nr_pages = $page_max{$field} || next;
510     #print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n";
511     #print STDERR Dumper($cache->{display_data});
512     for (my $page=0; $page <= $nr_pages; $page++) {
513 dpavlin 195 my $display_data;
514     if ($cache->{format}->{$field}) {
515 dpavlin 207 my $tmp = mkformat($cache->{format}->{$field},$cache->{display_data}->{$field}->[$page]);
516     $display_data=$tmp if ($tmp);
517 dpavlin 195 } else {
518     $display_data = $cache->{display_data}->{$field}->[$page];
519     }
520 dpavlin 178 if ($display_data) { # default
521     if ($field eq "headline") {
522     $xml .= xmlify("headline", $display_data);
523     } else {
524 dpavlin 29
525 dpavlin 178 # fallback to empty field name if needed
526     $html .= get_field_name($config,$field,$field_usage{$field}) || '';
527     $html .= "#-#".$display_data."###\n";
528     }
529     }
530    
531     my $swish_data = $cache->{swish_data}->{$field}->[$page];
532     if ($swish_data) {
533     # remove extra spaces
534     $swish_data =~ s/ +/ /g;
535     $swish_data =~ s/ +$//g;
536 dpavlin 13
537 dpavlin 320 $xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data));
538 dpavlin 13 }
539 dpavlin 178
540     my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page];
541     if ($swish_exact_data) {
542     $swish_exact_data =~ s/ +/ /g;
543     $swish_exact_data =~ s/ +$//g;
544    
545     # add delimiters before and after word.
546     # That is required to produce exact match
547 dpavlin 320 $xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data));
548 dpavlin 13 }
549 dpavlin 182
550 dpavlin 188 my $idel = $cache->{index_delimiter}->{$field};
551     foreach my $tmp (@{$cache->{index_data}->{$field}->[$page]}) {
552     my $i = $tmp;
553     my $d = $tmp;
554     if ($idel && $tmp =~ m/$idel/) {
555     ($i,$d) = split(/$idel/,$tmp);
556     }
557     $index->insert($field, $i, $d, $path);
558     #print STDERR "index [$idel] $field: $i --> $d [$path]\n";
559 dpavlin 182 }
560 dpavlin 13 }
561 dpavlin 178
562 dpavlin 10 }
563 dpavlin 178
564     if (! $iterate_by_page) {
565     if ($display_data) {
566     if ($field eq "headline") {
567     $xml .= xmlify("headline", $display_data);
568     } else {
569 dpavlin 20
570 dpavlin 178 # fallback to empty field name if needed
571     $html .= get_field_name($config,$field,$field_usage{$field}) || '';
572     $html .= "#-#".$display_data."###\n";
573     }
574     }
575     if ($swish_data) {
576     # remove extra spaces
577     $swish_data =~ s/ +/ /g;
578     $swish_data =~ s/ +$//g;
579 dpavlin 10
580 dpavlin 320 $xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data));
581 dpavlin 178 }
582 dpavlin 9
583 dpavlin 178 if ($swish_exact_data) {
584     $swish_exact_data =~ s/ +/ /g;
585     $swish_exact_data =~ s/ +$//g;
586    
587     # add delimiters before and after word.
588     # That is required to produce exact match
589 dpavlin 320 $xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data));
590 dpavlin 178 }
591 dpavlin 163 }
592 dpavlin 3 }
593 dpavlin 13
594     # dump formatted output in <html>
595     if ($html) {
596 dpavlin 81 #$xml .= xmlify("html",$html);
597     $xml .= "<html><![CDATA[ $html ]]></html>";
598 dpavlin 13 }
599    
600 dpavlin 3 if ($xml) {
601 dpavlin 13 $xml .= $add_xml if ($add_xml);
602 dpavlin 10 return "<xml>\n$xml</xml>\n";
603 dpavlin 3 } else {
604     return;
605     }
606     }
607    
608     ##########################################################################
609    
610 dpavlin 54 # read configuration for this script
611 dpavlin 13 my $cfg = new Config::IniFiles( -file => $config_file );
612 dpavlin 1
613 dpavlin 54 # read global.conf configuration
614     my $cfg_global = new Config::IniFiles( -file => 'global.conf' );
615    
616 dpavlin 50 # open index
617     $index = new index_DBI(
618 dpavlin 54 $cfg_global->val('global', 'dbi_dbd'),
619     $cfg_global->val('global', 'dbi_dsn'),
620     $cfg_global->val('global', 'dbi_user'),
621     $cfg_global->val('global', 'dbi_passwd') || '',
622 dpavlin 50 );
623    
624 dpavlin 97 my $show_progress = $cfg_global->val('global', 'show_progress');
625    
626 dpavlin 320 my $my_unac_filter = $cfg_global->val('global', 'my_unac_filter');
627     if ($my_unac_filter) {
628     print STDERR "using $my_unac_filter to filter characters for search\n";
629     require $my_unac_filter;
630     } else {
631     print STDERR "### fallback to default my_unac_string!\n";
632     eval q{
633     sub main::my_unac_string($$) {
634     my ($charset, $string) = (@_);
635     return $string;
636     }
637     };
638 dpavlin 164 }
639    
640 dpavlin 13 foreach my $database ($cfg->Sections) {
641 dpavlin 1
642 dpavlin 672 # save database name in global variable path for later
643     # (need for index filter creation)
644     $path = $database;
645    
646 dpavlin 54 my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined";
647 dpavlin 40 my $add_xml = $cfg -> val($database, 'xml'); # optional
648 dpavlin 1
649 dpavlin 177 # create new lookup file
650     my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional
651     if ($lookup_file) {
652     #tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644;
653 dpavlin 379 if (! -e $lookup_file) {
654     open(LOOKUP, "> $lookup_file") || die "can't create $lookup_file': $!";
655     close(LOOKUP);
656     }
657 dpavlin 177 tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644;
658     print STDERR "creating lookup file '$lookup_file'\n";
659 dpavlin 231 # delete memory cache for lookup file
660     delete $cache->{lhash};
661 dpavlin 177 }
662    
663     # open existing lookup file
664     $lookup_file = $cfg -> val($database, 'lookup_open'); # optional
665     if ($lookup_file) {
666     #tie %lhash, 'GDBM_File', $lookup_file, &GDBM_READER, 0644;
667     tie %lhash, 'TDB_File', $lookup_file, TDB_DEFAULT, O_RDWR, 0644;
668     print STDERR "opening lookup file '$lookup_file'\n";
669     }
670    
671 dpavlin 678 my $import_xml_file = "./import_xml/$type.xml";
672 dpavlin 1
673 dpavlin 678 if (! -r $import_xml_file) {
674     print STDERR "ERROR: file $import_xml_file not readable skipping!\n";
675     next;
676     }
677    
678     print STDERR "reading $import_xml_file\n";
679    
680 dpavlin 58 # extract just type basic
681     my $type_base = $type;
682     $type_base =~ s/_.+$//g;
683 dpavlin 40
684 dpavlin 678 $config=XMLin($import_xml_file, ForceArray => [ $type2tag{$type_base}, 'config', 'format' ], ForceContent => 1 );
685 dpavlin 58
686 dpavlin 618 # helper for progress bar
687     sub fmt_time {
688     my $t = shift || 0;
689     my $out = "";
690    
691     my ($ss,$mm,$hh) = gmtime($t);
692     $out .= "${hh}h" if ($hh);
693     $out .= sprintf("%02d:%02d", $mm,$ss);
694     $out .= " " if ($hh == 0);
695     return $out;
696     }
697    
698 dpavlin 54 # output current progress indicator
699     my $last_p = 0;
700 dpavlin 618 my $start_t = time();
701 dpavlin 54 sub progress {
702 dpavlin 101 return if (! $show_progress);
703 dpavlin 620 my $current = shift;
704 dpavlin 57 my $total = shift || 1;
705 dpavlin 54 my $p = int($current * 100 / $total);
706 dpavlin 628 if ($p < $last_p || $current == 1) {
707 dpavlin 626 $start_t = time();
708 dpavlin 628 $last_p = 0;
709 dpavlin 626 } elsif ($p != $last_p) {
710 dpavlin 618 my $rate = ($current / (time() - $start_t || 1));
711     my $eta = ($total-$current) / ($rate || 1);
712     printf STDERR ("%5d [%-38s] %-5d %0.1f/s %s\r",$current,"=" x ($p/3)."$p%>", $total, $rate, fmt_time($eta));
713 dpavlin 54 $last_p = $p;
714     }
715 dpavlin 3 }
716    
717 dpavlin 67 my $fake_dir = 1;
718 dpavlin 628 my $fake_pos = 0;
719     my $last_fake_t = time();
720 dpavlin 67 sub fakeprogress {
721 dpavlin 104 return if (! $show_progress);
722 dpavlin 67 my $current = shift @_;
723    
724 dpavlin 628 my @ind = ('-','\\','|','/','-','\\','|','/');
725 dpavlin 67
726 dpavlin 628 if ($current < $fake_pos) {
727     $start_t = time();
728     $last_fake_t = 0;
729     $fake_dir = 1;
730     $fake_pos = 0;
731 dpavlin 67 }
732 dpavlin 628
733     if (time()-$last_fake_t >= 1) {
734     $last_fake_t = time();
735     $fake_pos += $fake_dir;
736     $fake_dir = -$fake_dir if ($fake_pos > 38);
737     }
738    
739     if ($current % 10 == 0) {
740     my $rate = ($current / (time() - $start_t || 1));
741     printf STDERR ("%5d [%-38s] %0.1f/s\r",$current, " " x $fake_pos .$ind[($current / 10) % 8], $rate);
742     }
743 dpavlin 67 }
744    
745 dpavlin 54 # now read database
746     print STDERR "using: $type...\n";
747 dpavlin 1
748 dpavlin 180 # erase cache for tags by order in this database
749     delete $cache->{tags_by_order};
750    
751 dpavlin 58 if ($type_base eq "isis") {
752    
753 dpavlin 54 my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!";
754 dpavlin 1
755 dpavlin 54 $import2cp = Text::Iconv->new($config->{isis_codepage},$codepage);
756 dpavlin 632 my $db = new Biblio::Isis( isisdb => $isis_db );
757 dpavlin 10
758 dpavlin 673 if (! $db) {
759 dpavlin 650 print STDERR "FATAL: can't read ISIS database: $isis_db, skipping...\n";
760 dpavlin 647 next;
761     }
762    
763 dpavlin 673 my $max_rowid = $db->count if ($db);
764    
765 dpavlin 54 print STDERR "Reading database: $isis_db [$max_rowid rows]\n";
766    
767     for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) {
768 dpavlin 622 my $row = $db->to_hash( $row_id );
769     if ($row) {
770    
771     $row->{mfn} = $row_id;
772     $row->{record} = $db->{record};
773    
774 dpavlin 54 progress($row->{mfn}, $max_rowid);
775    
776     my $swishpath = $path."#".int($row->{mfn});
777    
778 dpavlin 59 if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) {
779 dpavlin 54 $xml = $cp2utf->convert($xml);
780     use bytes; # as opposed to chars
781     print "Path-Name: $swishpath\n";
782     print "Content-Length: ".(length($xml)+1)."\n";
783     print "Document-Type: XML\n\n$xml\n";
784     }
785 dpavlin 3 }
786 dpavlin 54 }
787     print STDERR "\n";
788 dpavlin 3
789 dpavlin 58 } elsif ($type_base eq "excel") {
790 dpavlin 255 require Spreadsheet::ParseExcel;
791     require Spreadsheet::ParseExcel::Utility;
792     import Spreadsheet::ParseExcel::Utility qw(int2col);
793 dpavlin 54
794     $import2cp = Text::Iconv->new($config->{excel_codepage},$codepage);
795     my $excel_file = $cfg -> val($database, 'excel_file') || die "$database doesn't have 'excel_file' defined!";
796 dpavlin 43
797 dpavlin 54 my $sheet = x($config->{sheet}) || die "no sheet in $type.xml";
798 dpavlin 56 my $start_row = x($config->{start_row}) - 1 || die "no start_row in $type.xml";
799 dpavlin 54
800     my $oBook = Spreadsheet::ParseExcel::Workbook->Parse($excel_file) || die "can't open Excel file '$excel_file'";
801    
802     my $sheet_nr = 0;
803     foreach my $oWks (@{$oBook->{Worksheet}}) {
804     #print STDERR "-- SHEET $sheet_nr:", $oWks->{Name}, "\n";
805     last if ($oWks->{Name} eq $sheet);
806     $sheet_nr++;
807     }
808    
809     my $oWorksheet = $oBook->{Worksheet}[$sheet_nr];
810    
811     print STDERR "using sheet: ",$oWorksheet->{Name},"\n";
812     defined ($oWorksheet) || die "can't find sheet '$sheet' in $excel_file";
813     my $end_row = x($config->{end_row}) || $oWorksheet->{MaxRow};
814    
815 dpavlin 56 for(my $iR = $start_row ; defined $end_row && $iR <= $end_row ; $iR++) {
816 dpavlin 54 my $row;
817     for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) {
818     my $cell = $oWorksheet->{Cells}[$iR][$iC];
819     if ($cell) {
820 dpavlin 488 # this conversion is a cludge.
821     # Files from Excell could have
822     # characters which don't fit into
823     # destination encoding.
824     $row->{int2col($iC)} = $utf2cp->convert($cell->Value) || $cell->Value;
825 dpavlin 54 }
826     }
827    
828     progress($iR, $end_row);
829    
830     # print "row[$iR/$end_row] ";
831     # foreach (keys %{$row}) {
832     # print "$_: ",$row->{$_},"\t";
833     # }
834     # print "\n";
835    
836     my $swishpath = $database."#".$iR;
837    
838     next if (! $row);
839    
840 dpavlin 59 if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) {
841 dpavlin 44 $xml = $cp2utf->convert($xml);
842 dpavlin 35 use bytes; # as opposed to chars
843 dpavlin 43 print "Path-Name: $swishpath\n";
844 dpavlin 10 print "Content-Length: ".(length($xml)+1)."\n";
845     print "Document-Type: XML\n\n$xml\n";
846 dpavlin 3 }
847 dpavlin 1 }
848 dpavlin 672
849     print STDERR "\n";
850    
851 dpavlin 62 } elsif ($type_base eq "marc") {
852 dpavlin 67
853 dpavlin 620 require MARC::File::USMARC;
854 dpavlin 62
855     $import2cp = Text::Iconv->new($config->{marc_codepage},$codepage);
856     my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!";
857    
858     # optional argument is format
859 dpavlin 620 warn "marc_format is no longer used!" if ($config->{marc_format});
860 dpavlin 62 print STDERR "Reading MARC file '$marc_file'\n";
861    
862 dpavlin 650 my $marc = MARC::File::USMARC->in( $marc_file );
863 dpavlin 62
864 dpavlin 650 if (! $marc) {
865     print STDERR "FATAL: can't read MARC file: $marc_file, skipping...\n";
866     next;
867     }
868    
869 dpavlin 620 # count records in MARC file
870     sub marc_count {
871     my $filename = shift || die;
872 dpavlin 650 my $file = MARC::File::USMARC->in($filename) || return;
873 dpavlin 620 my $count = 0;
874     while ($file->skip()) {
875     $count++;
876     }
877     return $count;
878     }
879 dpavlin 62
880 dpavlin 620 my $count = marc_count($marc_file) || warn "no records in '$marc_file'?";
881 dpavlin 62
882 dpavlin 650 my $i = 1;
883 dpavlin 62
884 dpavlin 620 while( my $rec = $marc->next() ) {
885 dpavlin 62
886 dpavlin 650 progress($i,$count);
887 dpavlin 215
888 dpavlin 62 my $swishpath = $database."#".$i;
889    
890 dpavlin 620 if (my $xml = data2xml($type_base,$rec,$add_xml,$cfg,$database)) {
891 dpavlin 62 $xml = $cp2utf->convert($xml);
892     use bytes; # as opposed to chars
893     print "Path-Name: $swishpath\n";
894     print "Content-Length: ".(length($xml)+1)."\n";
895     print "Document-Type: XML\n\n$xml\n";
896     }
897 dpavlin 650
898     $i++;
899 dpavlin 62 }
900 dpavlin 215
901     print STDERR "\n";
902    
903 dpavlin 67 } elsif ($type_base eq "feed") {
904    
905     $import2cp = Text::Iconv->new($config->{feed_codepage},$codepage);
906     my $prog = x($config->{prog}) || die "$database doesn't have 'prog' defined!";
907    
908     print STDERR "Reading feed from program '$prog'\n";
909    
910     open(FEED,"feeds/$prog |") || die "can't start $prog: $!";
911    
912 dpavlin 74 my $i=1; # record nr.
913 dpavlin 67
914     my $data;
915     my $line=1;
916    
917     while (<FEED>) {
918     chomp;
919    
920     if (/^$/) {
921 dpavlin 74 my $swishpath = $database."#".$i++;
922 dpavlin 67
923     if (my $xml = data2xml($type_base,$data,$add_xml,$cfg,$database)) {
924     $xml = $cp2utf->convert($xml);
925     use bytes; # as opposed to chars
926     print "Path-Name: $swishpath\n";
927     print "Content-Length: ".(length($xml)+1)."\n";
928     print "Document-Type: XML\n\n$xml\n";
929     }
930     $line = 1;
931     $data = {};
932     next;
933     }
934    
935 dpavlin 74 $line = $1 if (s/^(\d+):\s*//);
936 dpavlin 67 $data->{$line++} = $_;
937    
938 dpavlin 74 fakeprogress($i);
939 dpavlin 67
940     }
941 dpavlin 177 # close lookup
942     untie %lhash if (%lhash);
943 dpavlin 678
944     } elsif ($type_base eq "dbf") {
945    
946     my $dbf_file = $cfg -> val($database, 'dbf_file') || die "$database doesn't have 'dbf_file' defined!";
947     my $dbf_codepage = $cfg -> val($database, 'dbf_codepage') || die "$database doesn't have 'dbf_codepage' defined!";
948     my $dbf_mapping = $cfg -> val($database, 'dbf_mapping') || die "$database doesn't have 'dbf_mapping' defined!";
949    
950     $import2cp = Text::Iconv->new($dbf_codepage,$codepage);
951     require XBase;
952     my $db = new XBase $dbf_file;
953    
954     if (! $db) {
955     print STDERR "ERROR: can't read DBF database: $dbf_file, skipping...\n";
956     next;
957     }
958    
959     my $max_rowid = $db->last_record;
960    
961     print STDERR "Reading database: $dbf_file [$max_rowid rows]\n";
962    
963     my %dbf2iso;
964     foreach my $m (split(/[\n\r]+/,$dbf_mapping)) {
965     my ($col,$fld) = split(/\s+/,$m,2);
966     $dbf2iso{$col} = $fld;
967     }
968    
969     #print STDERR "## dbf2iso: ",Dumper(\%dbf2iso),"\n## /dbf2iso\n";
970    
971     # bad, bad...
972     require "to_hash.pm";
973    
974     foreach my $row_id (0 .. $max_rowid) {
975     my $dbf_row = $db->get_record_as_hash($row_id);
976     if ($dbf_row) {
977    
978     #print STDERR "## dbf_row: ",Dumper($dbf_row),"\n## /dbf_row\n";
979     # apply mapping from config file
980     # all unspecified records will get _ in
981     # front of them - _DELETE will be __DELETE
982     my $rec;
983     map {
984     my $new_fld = $dbf2iso{$_} || '_'.$_;
985     my $data = $dbf_row->{$_};
986     push @{ $rec->{$new_fld} }, $data if ($data && $data !~ /^(?:\s+|\$a\.|)$/);
987     } keys %{$dbf_row};
988     #print STDERR "## rec: ",Dumper($rec),"\n## /rec\n";
989     my $row = to_hash($row_id+1, $rec);
990    
991     $row->{mfn} = $row_id+1;
992     $row->{record} = $rec;
993    
994     #print STDERR "## row: ",Dumper($row),"\n## /row\n";
995     progress($row->{mfn}, $max_rowid);
996    
997     my $swishpath = $path."#".int($row->{mfn});
998    
999     if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) {
1000     $xml = $cp2utf->convert($xml);
1001     use bytes; # as opposed to chars
1002     print "Path-Name: $swishpath\n";
1003     print "Content-Length: ".(length($xml)+1)."\n";
1004     print "Document-Type: XML\n\n$xml\n";
1005     }
1006     }
1007     }
1008     print STDERR "\n";
1009 dpavlin 1 }
1010     }
1011 dpavlin 3
1012 dpavlin 10 # call this to commit index
1013     $index->close;
1014 dpavlin 3
1015     1;
1016     __END__
1017     ##########################################################################
1018    
1019     =head1 NAME
1020    
1021 dpavlin 81 all2xml.pl - read various file formats and dump XML for SWISH-E
1022 dpavlin 3
1023 dpavlin 298 =head1 SYNOPSYS
1024    
1025     $ all2xml.pl [test.conf]
1026    
1027 dpavlin 3 =head1 DESCRIPTION
1028    
1029 dpavlin 650 This command will read ISIS data file using Biblio::Isis perl module, MARC
1030     records using MARC::File module and optionally Micro$oft Excel files to
1031 dpavlin 81 create one XML file for usage with I<SWISH-E> indexer. Dispite it's name,
1032     this script B<isn't general xml generator> from isis files (isis allready
1033     has something like that). Output of this script is tailor-made for SWISH-E.
1034 dpavlin 3
1035 dpavlin 298 If no configuration file is specified, it will use default one called
1036     C<all2xml.conf>.
1037    
1038 dpavlin 81 =head1 BUGS
1039    
1040     Documentation is really lacking. However, in true Open Source spirit, source
1041     is best documentation. I even made considerable effort to comment parts
1042     which are not intuitively clear, so...
1043    
1044 dpavlin 3 =head1 AUTHOR
1045    
1046     Dobrica Pavlinusic <dpavlin@rot13.org>
1047    
1048     =head1 COPYRIGHT
1049    
1050     GNU Public License (GPL) v2 or later
1051    
1052     =head1 SEE ALSO
1053    
1054     SWISH-E web site at http://www.swish-e.org
1055    
1056     =cut

Properties

Name Value
cvs2svn:cvs-rev 1.64
svn:executable *
  ViewVC Help
Powered by ViewVC 1.1.26